from sklearn_benchmarks.reporting.hp_match import HPMatchReporting

scikit-learn vs. ONNX Runtime

reporting = HPMatchReporting(against_lib="onnx", config="config.yml")
reporting.make_report()

We assume here there is a perfect match between the hyperparameters of both librairies. For a given set of parameters and a given dataset, we compute the speedup time scikit-learn / time onnx. For instance, a speedup of 2 means that onnx is twice as fast as scikit-learn for a given set of parameters and a given dataset.

KNeighborsClassifier_brute_force

onnx (1.10.1) vs. scikit-learn (1.0.dev0)

Speedup barplots

All estimators share the following parameters: algorithm=brute.

WARNING! Mismatch between validation scores for 12 predictions (75.0%). See below.

	estimator	function	n_samples_train	n_samples	n_features	accuracy_score_sklearn	accuracy_score_onnx	diff_accuracy_scores
1	KNeighborsClassifier_brute_force	predict	100000	1000	100	0.663	1.000	0.337
2	KNeighborsClassifier_brute_force	predict	100000	1	100	1.000	0.757	0.243
4	KNeighborsClassifier_brute_force	predict	100000	1000	100	0.757	0.882	0.125
7	KNeighborsClassifier_brute_force	predict	100000	1000	100	0.882	1.000	0.118
8	KNeighborsClassifier_brute_force	predict	100000	1	100	1.000	0.757	0.243
10	KNeighborsClassifier_brute_force	predict	100000	1000	100	0.882	0.663	0.219
13	KNeighborsClassifier_brute_force	predict	100000	1000	100	0.757	1.000	0.243
14	KNeighborsClassifier_brute_force	predict	100000	1	100	1.000	0.922	0.078
16	KNeighborsClassifier_brute_force	predict	100000	1000	100	0.663	0.929	0.266
19	KNeighborsClassifier_brute_force	predict	100000	1000	2	0.896	1.000	0.104
20	KNeighborsClassifier_brute_force	predict	100000	1	2	1.000	0.922	0.078
22	KNeighborsClassifier_brute_force	predict	100000	1000	2	0.922	0.896	0.026

Raw results

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_jobs	n_neighbors	mean_duration_onnx	std_duration_onnx	accuracy_score_onnx	speedup	std_speedup	sklearn_profiling	onnx_profiling
0	KNeighborsClassifier_brute_force	fit	100000	100000	100	0.012	0.0	6.729	0.0	-1	1	17.049	0.123	0.663	0.001	0.001	See	See
3	KNeighborsClassifier_brute_force	fit	100000	100000	100	0.012	0.0	6.908	0.0	-1	5	0.316	0.004	1.000	0.037	0.037	See	See
6	KNeighborsClassifier_brute_force	fit	100000	100000	100	0.012	0.0	6.794	0.0	1	100	17.214	0.019	0.882	0.001	0.001	See	See
9	KNeighborsClassifier_brute_force	fit	100000	100000	100	0.012	0.0	6.831	0.0	-1	100	0.317	0.006	1.000	0.037	0.037	See	See
12	KNeighborsClassifier_brute_force	fit	100000	100000	100	0.012	0.0	6.884	0.0	1	5	3.769	0.010	0.896	0.003	0.003	See	See
15	KNeighborsClassifier_brute_force	fit	100000	100000	100	0.012	0.0	6.663	0.0	1	1	0.239	0.003	1.000	0.050	0.050	See	See
18	KNeighborsClassifier_brute_force	fit	100000	100000	2	0.004	0.0	0.377	0.0	-1	1	3.750	0.008	0.929	0.001	0.001	See	See
21	KNeighborsClassifier_brute_force	fit	100000	100000	2	0.004	0.0	0.379	0.0	-1	5	0.240	0.004	1.000	0.018	0.018	See	See

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_jobs	n_neighbors	mean_duration_onnx	std_duration_onnx	accuracy_score_onnx	speedup	std_speedup	sklearn_profiling	onnx_profiling
1	KNeighborsClassifier_brute_force	predict	100000	1000	100	1.969	0.081	0.000	0.002	-1	1	0.315	0.004	1.000	6.249	6.250	See	See
2	KNeighborsClassifier_brute_force	predict	100000	1	100	0.023	0.002	0.000	0.023	-1	1	17.135	0.199	0.757	0.001	0.001	See	See
4	KNeighborsClassifier_brute_force	predict	100000	1000	100	2.849	0.043	0.000	0.003	-1	5	16.983	0.068	0.882	0.168	0.168	See	See
5	KNeighborsClassifier_brute_force	predict	100000	1	100	0.025	0.002	0.000	0.025	-1	5	0.317	0.007	1.000	0.078	0.078	See	See
7	KNeighborsClassifier_brute_force	predict	100000	1000	100	2.051	0.041	0.000	0.002	1	100	0.317	0.004	1.000	6.479	6.479	See	See
8	KNeighborsClassifier_brute_force	predict	100000	1	100	0.025	0.004	0.000	0.025	1	100	17.162	0.152	0.757	0.001	0.001	See	See
10	KNeighborsClassifier_brute_force	predict	100000	1000	100	2.849	0.052	0.000	0.003	-1	100	17.359	0.024	0.663	0.164	0.164	See	See
11	KNeighborsClassifier_brute_force	predict	100000	1	100	0.025	0.003	0.000	0.025	-1	100	0.307	0.006	1.000	0.082	0.082	See	See
13	KNeighborsClassifier_brute_force	predict	100000	1000	100	2.025	0.003	0.000	0.002	1	5	0.239	0.004	1.000	8.470	8.471	See	See
14	KNeighborsClassifier_brute_force	predict	100000	1	100	0.022	0.000	0.000	0.022	1	5	3.711	0.010	0.922	0.006	0.006	See	See
16	KNeighborsClassifier_brute_force	predict	100000	1000	100	1.177	0.012	0.001	0.001	1	1	3.753	0.009	0.929	0.314	0.314	See	See
17	KNeighborsClassifier_brute_force	predict	100000	1	100	0.021	0.000	0.000	0.021	1	1	0.239	0.005	1.000	0.086	0.086	See	See
19	KNeighborsClassifier_brute_force	predict	100000	1000	2	1.806	0.027	0.000	0.002	-1	1	0.240	0.004	1.000	7.528	7.529	See	See
20	KNeighborsClassifier_brute_force	predict	100000	1	2	0.006	0.005	0.000	0.006	-1	1	3.720	0.008	0.922	0.002	0.002	See	See
22	KNeighborsClassifier_brute_force	predict	100000	1000	2	2.691	0.023	0.000	0.003	-1	5	3.764	0.006	0.896	0.715	0.715	See	See
23	KNeighborsClassifier_brute_force	predict	100000	1	2	0.006	0.003	0.000	0.006	-1	5	0.239	0.005	1.000	0.027	0.027	See	See

KNeighborsClassifier_kd_tree

onnx (1.10.1) vs. scikit-learn (1.0.dev0)

Speedup barplots

All estimators share the following parameters: algorithm=kd_tree.

WARNING! Mismatch between validation scores for 12 predictions (75.0%). See below.

	estimator	function	n_samples_train	n_samples	n_features	accuracy_score_sklearn	accuracy_score_onnx	diff_accuracy_scores
1	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	0.929	1.000	0.071
2	KNeighborsClassifier_kd_tree	predict	1000000	1	10	1.000	0.946	0.054
4	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	0.946	0.951	0.005
7	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	0.951	1.000	0.049
8	KNeighborsClassifier_kd_tree	predict	1000000	1	10	1.000	0.946	0.054
10	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	0.951	0.929	0.022
13	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	0.946	1.000	0.054
14	KNeighborsClassifier_kd_tree	predict	1000000	1	10	1.000	0.911	0.089
16	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	0.929	0.894	0.035
19	KNeighborsClassifier_kd_tree	predict	1000	1000	2	0.891	1.000	0.109
20	KNeighborsClassifier_kd_tree	predict	1000	1	2	1.000	0.911	0.089
22	KNeighborsClassifier_kd_tree	predict	1000	1000	2	0.911	0.891	0.020

Raw results

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_jobs	n_neighbors	mean_duration_onnx	std_duration_onnx	accuracy_score_onnx	speedup	std_speedup	sklearn_profiling	onnx_profiling
0	KNeighborsClassifier_kd_tree	fit	1000000	1000000	10	3.191	0.145	0.025	0.0	-1	1	107.463	0.00	0.929	0.030	0.030	See	See
3	KNeighborsClassifier_kd_tree	fit	1000000	1000000	10	3.730	0.104	0.021	0.0	-1	5	2.605	0.16	1.000	1.432	1.435	See	See
6	KNeighborsClassifier_kd_tree	fit	1000000	1000000	10	3.731	0.124	0.021	0.0	1	100	104.490	0.00	0.951	0.036	0.036	See	See
9	KNeighborsClassifier_kd_tree	fit	1000000	1000000	10	4.036	0.055	0.020	0.0	-1	100	2.621	0.15	1.000	1.540	1.542	See	See
12	KNeighborsClassifier_kd_tree	fit	1000000	1000000	10	3.750	0.041	0.021	0.0	1	5	0.038	0.00	0.891	98.562	98.565	See	See
15	KNeighborsClassifier_kd_tree	fit	1000000	1000000	10	3.710	0.037	0.022	0.0	1	1	0.005	0.00	1.000	727.161	727.277	See	See
18	KNeighborsClassifier_kd_tree	fit	1000	1000	2	0.001	0.001	0.020	0.0	-1	1	0.060	0.00	0.894	0.013	0.013	See	See
21	KNeighborsClassifier_kd_tree	fit	1000	1000	2	0.001	0.000	0.031	0.0	-1	5	0.005	0.00	1.000	0.101	0.101	See	See

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	iteration_throughput	latency	n_jobs	n_neighbors	mean_duration_onnx	std_duration_onnx	accuracy_score_onnx	speedup	std_speedup	sklearn_profiling	onnx_profiling
1	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	0.818	0.964	0.000	0.001	-1	1	2.612	0.197	1.000	0.313	0.314	See	See
2	KNeighborsClassifier_kd_tree	predict	1000000	1	10	0.002	0.001	0.000	0.002	-1	1	105.443	0.000	0.946	0.000	0.000	See	See
4	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	1.013	0.282	0.000	0.001	-1	5	106.188	0.000	0.951	0.010	0.010	See	See
5	KNeighborsClassifier_kd_tree	predict	1000000	1	10	0.003	0.001	0.000	0.003	-1	5	2.587	0.116	1.000	0.001	0.001	See	See
7	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	5.678	0.240	0.000	0.006	1	100	2.616	0.136	1.000	2.171	2.173	See	See
8	KNeighborsClassifier_kd_tree	predict	1000000	1	10	0.004	0.001	0.000	0.004	1	100	104.729	0.000	0.946	0.000	0.000	See	See
10	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	3.289	0.184	0.000	0.003	-1	100	106.575	0.000	0.929	0.031	0.031	See	See
11	KNeighborsClassifier_kd_tree	predict	1000000	1	10	0.005	0.001	0.000	0.005	-1	100	2.601	0.136	1.000	0.002	0.002	See	See
13	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	1.652	0.158	0.000	0.002	1	5	0.005	0.000	1.000	328.204	328.280	See	See
14	KNeighborsClassifier_kd_tree	predict	1000000	1	10	0.002	0.001	0.000	0.002	1	5	0.040	0.001	0.911	0.038	0.038	See	See
16	KNeighborsClassifier_kd_tree	predict	1000000	1000	10	0.925	0.191	0.000	0.001	1	1	0.060	0.002	0.894	15.372	15.378	See	See
17	KNeighborsClassifier_kd_tree	predict	1000000	1	10	0.001	0.000	0.000	0.001	1	1	0.005	0.000	1.000	0.201	0.201	See	See
19	KNeighborsClassifier_kd_tree	predict	1000	1000	2	0.026	0.012	0.001	0.000	-1	1	0.005	0.000	1.000	5.108	5.109	See	See
20	KNeighborsClassifier_kd_tree	predict	1000	1	2	0.002	0.000	0.000	0.002	-1	1	0.039	0.000	0.911	0.051	0.051	See	See
22	KNeighborsClassifier_kd_tree	predict	1000	1000	2	0.023	0.002	0.001	0.000	-1	5	0.039	0.001	0.891	0.589	0.589	See	See
23	KNeighborsClassifier_kd_tree	predict	1000	1	2	0.002	0.000	0.000	0.002	-1	5	0.005	0.000	1.000	0.377	0.377	See	See

HistGradientBoostingClassifier_best

onnx (1.10.1) vs. scikit-learn (1.0.dev0)

Speedup barplots

All estimators share the following parameters: learning_rate=0.01, n_iter_no_change=10.0, max_leaf_nodes=100.0, max_bins=255.0, min_samples_leaf=100.0, max_iter=300.0.

WARNING! Mismatch between validation scores for 1 prediction (100.0%). See below.

	estimator	function	n_samples_train	n_samples	n_features	accuracy_score_sklearn	accuracy_score_onnx	diff_accuracy_scores
1	HistGradientBoostingClassifier_best	predict	100000	1000	100	0.824	1.0	0.176

Raw results

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	mean_duration_onnx	std_duration_onnx	accuracy_score_onnx	speedup	std_speedup	sklearn_profiling	onnx_profiling
0	HistGradientBoostingClassifier_best	fit	100000	100000	100	97.843	0.0	300	0.001	0.001	0.504	0.02	0.824	194.184	194.343	See	See

	estimator	function	n_samples_train	n_samples	n_features	mean_duration_sklearn	std_duration_sklearn	n_iter	iteration_throughput	latency	mean_duration_onnx	std_duration_onnx	accuracy_score_onnx	speedup	std_speedup	sklearn_profiling	onnx_profiling
1	HistGradientBoostingClassifier_best	predict	100000	1000	100	0.114	0.002	300	0.007	0.0	0.421	0.01	1.0	0.27	0.27	See	See